Imports

from utils import print_time_report, print_results, plot_results

Time report

print_time_report()

	hour	min	sec
algo
KNeighborsClassifier	0	13	23
daal4py_KNeighborsClassifier	0	3	15
KNeighborsClassifier_kd_tree	0	5	41
daal4py_KNeighborsClassifier_kd_tree	0	1	30
KMeans	0	19	27
daal4py_KMeans	0	6	15
total	0	49	34

KNeighborsClassifier - Brute Force

print_results(algo="KNeighborsClassifier", versus_lib="daal4py")

plot_results(algo="KNeighborsClassifier", versus_lib="daal4py", group_by_cols=["algorithm", "n_neighbors", "function"], split_hist_by=["n_jobs"])

KNeighborsClassifier - KD Tree

print_results(algo="KNeighborsClassifier_kd_tree", versus_lib="daal4py")

plot_results(algo="KNeighborsClassifier_kd_tree", versus_lib="daal4py", group_by_cols=["algorithm", "n_neighbors", "function"], split_hist_by=["n_jobs"])

KMeans

from utils import _make_dataset
data = _make_dataset('KMeans', 'daal4py', compare_cols=["n_iter"])

data[data["function"] == "fit"].sort_values(["n_iter_sklearn", "n_iter_daal4py"])

	estimator	lib	function	n_samples	n_features	init	max_iter	n_clusters	n_init	tol	adjusted_rand_score	mean_sklearn	stdev_sklearn	n_iter_sklearn	mean_daal4py	stdev_daal4py	n_iter_daal4py	speedup	stdev_speedup
0	KMeans	sklearn	fit	10000	2	k-means++	30	3	1	0.0	NaN	0.0431	0.0438	19.0	0.0043	0.0010	18.0	10.02	43.80
9	KMeans	sklearn	fit	10000	2	random	30	3	1	0.0	NaN	0.0846	0.0333	22.0	0.0071	0.0007	30.0	11.92	47.57
18	KMeans	sklearn	fit	10000	100	k-means++	30	3	1	0.0	NaN	0.4918	0.0203	30.0	0.1087	0.0070	30.0	4.52	2.90
27	KMeans	sklearn	fit	10000	100	random	30	3	1	0.0	NaN	0.0161	0.0002	30.0	0.0054	0.0006	30.0	2.98	0.33
36	KMeans	sklearn	fit	1000000	2	k-means++	30	3	1	0.0	NaN	0.0257	0.0006	30.0	0.0056	0.0005	30.0	4.59	1.20
45	KMeans	sklearn	fit	1000000	2	random	30	3	1	0.0	NaN	0.3649	0.0767	30.0	0.0528	0.0013	30.0	6.91	59.00
54	KMeans	sklearn	fit	1000000	100	k-means++	30	3	1	0.0	NaN	0.1629	0.0072	30.0	0.0145	0.0008	30.0	11.23	9.00
63	KMeans	sklearn	fit	1000000	100	random	30	3	1	0.0	NaN	0.2086	0.0048	30.0	0.0285	0.0026	30.0	7.32	1.85
3	KMeans	sklearn	fit	10000	2	k-means++	30	10	1	0.0	NaN	0.0003	0.0001	NaN	0.0002	0.0001	NaN	1.50	1.00
6	KMeans	sklearn	fit	10000	2	k-means++	30	300	1	0.0	NaN	0.0003	0.0001	NaN	0.0001	0.0001	NaN	3.00	1.00
12	KMeans	sklearn	fit	10000	2	random	30	10	1	0.0	NaN	0.0004	0.0002	NaN	0.0002	0.0001	NaN	2.00	2.00
15	KMeans	sklearn	fit	10000	2	random	30	300	1	0.0	NaN	0.0003	0.0001	NaN	0.0001	0.0001	NaN	3.00	1.00
21	KMeans	sklearn	fit	10000	100	k-means++	30	10	1	0.0	NaN	0.0009	0.0001	NaN	0.0008	0.0002	NaN	1.12	0.50
24	KMeans	sklearn	fit	10000	100	k-means++	30	300	1	0.0	NaN	0.0003	0.0001	NaN	0.0002	0.0001	NaN	1.50	1.00
30	KMeans	sklearn	fit	10000	100	random	30	10	1	0.0	NaN	0.0004	0.0002	NaN	0.0002	0.0001	NaN	2.00	2.00
33	KMeans	sklearn	fit	10000	100	random	30	300	1	0.0	NaN	0.0003	0.0001	NaN	0.0001	0.0001	NaN	3.00	1.00
39	KMeans	sklearn	fit	1000000	2	k-means++	30	10	1	0.0	NaN	0.0004	0.0001	NaN	0.0002	0.0001	NaN	2.00	1.00
42	KMeans	sklearn	fit	1000000	2	k-means++	30	300	1	0.0	NaN	0.0003	0.0001	NaN	0.0001	0.0001	NaN	3.00	1.00
48	KMeans	sklearn	fit	1000000	2	random	30	10	1	0.0	NaN	0.0010	0.0003	NaN	0.0009	0.0001	NaN	1.11	3.00
51	KMeans	sklearn	fit	1000000	2	random	30	300	1	0.0	NaN	0.0003	0.0001	NaN	0.0001	0.0001	NaN	3.00	1.00
57	KMeans	sklearn	fit	1000000	100	k-means++	30	10	1	0.0	NaN	0.0007	0.0002	NaN	0.0003	0.0001	NaN	2.33	2.00
60	KMeans	sklearn	fit	1000000	100	k-means++	30	300	1	0.0	NaN	0.0003	0.0001	NaN	0.0002	0.0001	NaN	1.50	1.00
66	KMeans	sklearn	fit	1000000	100	random	30	10	1	0.0	NaN	0.0007	0.0002	NaN	0.0004	0.0001	NaN	1.75	2.00
69	KMeans	sklearn	fit	1000000	100	random	30	300	1	0.0	NaN	0.0003	0.0001	NaN	0.0002	0.0001	NaN	1.50	1.00

print_results(algo="KMeans", versus_lib="daal4py", compare_cols=["n_iter"])

plot_results(algo="KMeans", versus_lib="daal4py", group_by_cols=["init", "max_iter", "n_clusters", "n_init", "tol", "function"], compare_cols=["n_iter"])